Configural Frequency Analysis

Eye, Alexander von

doi:10.1002/0470013192.bsa129

Cited by 4 publications

(4 citation statements)

References 19 publications

(11 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…If these measures suggest that the model must be rejected, cell‐wise analysis of model‐data discrepancies will show the researcher where the model fails in particular. In some instances, that is, for specific log‐linear models, large cell‐specific discrepancies come with substantive interpretations, as has been discussed in the context of configural frequency analysis 17,18. Among the most frequently used residual measures are Pearson's and its standardized version, where i goes over the cells of the cross‐classification, and h i is the leverage which is the i th diagonal entry of the hat matrix, For Poisson GLMs, the diagonal entries, w ii , of matrix W ( W is a diagonal matrix) are the model frequencies \documentclass{article}\usepackage{amsmath}\usepackage{amssymb}\usepackage{amsbsy}\usepackage{amsfonts}\pagestyle{empty}\begin{document}$\hat{m}$\end{document} i .…”

Section: Residual Analysismentioning

confidence: 99%

Log‐linear modeling

Eye

Mun

Mair

2011

WIREs Computational Stats

Self Cite

View full text Add to dashboard Cite

This article describes log‐linear models as special cases of generalized linear models. Specifically, log‐linear models use a logarithmic link function. Log‐linear models are used to examine joint distributions of categorical variables, dependency relations, and association patterns. Three types of log‐linear models are discussed, hierarchical models, nonhierarchical models, and nonstandard models. Emphasis is placed on parameter interpretation. It is demonstrated that parameters are best interpretable when they represent the effects specified in the design matrix of the model. Parameter interpretation is illustrated first for a standard hierarchical model, and then for a nonstandard model that includes structural zeros. In a data example, the relationships among race of defendant, race of victim, and death penalty sentence are examined using a log‐linear model with all three two‐way interactions. Recent developments in log‐linear modeling are discussed. WIREs Comput Stat 2012, 4:218–223. doi: 10.1002/wics.203 This article is categorized under: Statistical Models > Generalized Linear Models

show abstract

Section: Residual Analysismentioning

confidence: 99%

Log‐linear modeling

Eye

Mun

Mair

2011

WIREs Computational Stats

Self Cite

View full text Add to dashboard Cite

show abstract

“…The local tests were post hoc tests and were described under the designation residual test or configuration frequency analysis (Von Eye et al. ; Stemmler ). We performed all analyses with the “cfa” package (Mair and Funke ).…”

Section: Methodsmentioning

confidence: 99%

Snorkeling‐Based Fish Stock Assessment by Anglers—A Valuable Method for Managing Recreational Fisheries

Pinter

Lautsch

Unfer

et al. 2018

N American J Fish Manag

View full text Add to dashboard Cite

Periodic assessment of harvested fish populations is essential for their sustainable management. A potential alternative to costly and resource‐intensive electrofishing estimates in clearwater streams is the noninvasive snorkeling method. To assess the utility of snorkeling for the angling community, we compared underwater fish counts carried out by novice snorkelers to state‐of‐the‐art electrofishing depletion estimates. Over two consecutive years, we sampled subadult and adult Brown Trout Salmo trutta and Rainbow Trout Oncorhynchus mykiss with both methods in a fourth‐order mountain stream. In each year, a new team of novice snorkelers collected the data. In total, 12 riffle, pool, and run habitats were sampled, and the homogeneity of abundance and size‐class distribution between the two methods was analyzed. Over both years, we could detect differences in 6 of 24 habitat × species configurations and in 10 of 72 habitat × species × size‐class configurations. Species‐specific behavioral traits and differences in the physical character of the habitats were responsible for a divergence in performance between the two methods. Overall, the observed effects were statistically interpreted as weak, as shown by local tests and the indicated low effect sizes. Snorkeling efficiency, however, remained affected by the effort and abilities of the team, as indicated by the year‐by‐year comparison. We conclude that in clearwater trout streams, snorkeling can be an appropriate substitute method for the widespread, autonomously organized angling community to gather data and build a sound foundation for fisheries‐related decision making, if limitations are considered.

show abstract

“…None of these, however, qualifies as a CFA base model (cf. the criteria for CFA base models listed above; von Eye, 2004). Therefore, we only discuss these three models.…”

Section: Statistically Comparing Lag Patternsmentioning

confidence: 99%

A Configural Perspective of Interindividual Differences in Intraindividual Change

Eye

Wiedermann

2016

JPOR

Self Cite

View full text Add to dashboard Cite

Lag analysis can be used to inspect stability and change of behavior over a pre-determined time interval, the lag. In the analysis of metric variables, lag analysis is well known and used to identify such temporal effects as seasonal trends. In the analysis of categorical variables, the same can be done. Either approach can be employed in the analysis of both aggregated and individual data. In the domain of studying individual cells of contingency tables, that is, in configural analysis, only two sources exist in which lag analysis is discussed (von Eye, Mair, & Mun, 2010;von Eye & Mun, 2012). In this paper, we place the method of configural lag analysis in a person-oriented context and propose new variants for the comparison of individuals. Three approaches are considered. The first involves searching for configural types and antitypes separately for the comparison individuals. The second approach can be viewed parallel to two-or multiple group Configural Frequency Analysis. Both approaches are presented within a log-linear framework. Configural base models are specified for the original configural lag method as well as the extended comparative methods, and questions are defined that can be answered using configural lag analysis. The third approach allows researchers to test hypotheses concerning groups of cells. In an empirical example, data are analyzed from a study on the development of drinking behavior in alcoholics. Further extensions and alternative methods of analysis are discussed.

show abstract

Configural Frequency Analysis

Cited by 4 publications

References 19 publications

Log‐linear modeling

Log‐linear modeling

Snorkeling‐Based Fish Stock Assessment by Anglers—A Valuable Method for Managing Recreational Fisheries

A Configural Perspective of Interindividual Differences in Intraindividual Change

Contact Info

Product

Resources

About