Figure 1 - uploaded by Micheline Manseau
Content may be subject to copyright.
![Three-dimensional plots of the: (A) estimated exchangeable, and (B) unstructured variance-covariance matrices estimated using the covariance pattern model for modeling long-term data from a forest fragmentation study [25]](profile/Micheline-Manseau/publication/27477613/figure/fig1/AS:669555823230982@1536645956559/Three-dimensional-plots-of-the-A-estimated-exchangeable-and-B-unstructured.png)
Three-dimensional plots of the: (A) estimated exchangeable, and (B) unstructured variance-covariance matrices estimated using the covariance pattern model for modeling long-term data from a forest fragmentation study [25]
Source publication
Ecological data sets often use clustered sampling, or use repeated sampling in a longitudinal design. Choosing the correct covariance structure is an important step in the analysis of such data, as the covariance dictates the degree of similarity among repeated observations. Three methods for choosing the covariance are: Akaike’s information criter...
Contexts in source publication
Context 1
... unstructured and exchangeable variance-covariance matrices esti- mated using the covariance pattern model are shown in Figure 1. The x- and y-axes show the years 1993 to 2007 and the z-axis shows the covariances among responses at the same site but at different years. ...
Context 2
... AIC and DIC both selected the unstructured covariance with the ex- changeable correlation as second best, which appears reasonable based on the three-dimensional plot of the estimated unstructured covariance (Fig- ure 1). The estimated number of parameters used by the DIC and AIC agreed closely, while the number of parameters from the trace used by the QIC was much smaller. ...
Similar publications
This work aims to evaluate which landscape features could influence the maturity of fragments of the Serra do Mar Coastal Forest in the southeastern Brazil. We measured vegetation structure of 11 forest fragments, and used basal area as indicator of forest maturity state. Landscape features were characteristics of forest fragment, including size, s...
This work aims to evaluate which landscape features could influence the maturity of fragments of the Serra do Mar Coastal Forest in the southeastern Brazil. We measured vegetation structure of 11 forest fragments, and used basal area as indicator of forest maturity state. Landscape features were characteristics of forest fragment, including size, s...
Citations
... Targets were famous faces from a wide range of occupations. While famous faces are not usually involved in forensic practice, there is evidence that they produce composites with correct naming that is similar to non-famous stimuli in composite research (e.g., Brace, Pike, & Kemp, 2000;Bruce, Ness, Hancock, Newman, & Rarity, 2002;Frowd, Bruce, Ness et al., 2007;Frowd et al., 2015;Frowd, Jones et al., 2013). Since we planned to select targets by attractiveness category, while controlling for other attributes of the face (memorability, likeability and familiarity), sampling stimuli from a large pool size was necessary. ...
... Individual rating scores (N = 1,152) were analysed using Generalized Estimating Equations (GEE), a regression-type approach that is statistically more powerful than ANOVA and provides a combined by-subjects and by-items model appropriate for the repeated observations of four sets of Likert ratings for each target photograph (Barnett, Koper, Dobson, Schmiegelow, & Manseau, 2009). The model used was full-factorial with an ordinal logistic response function and, based on homogenous correlations between elements, an exchangeable working correlation matrix. ...
... Correct naming of complete composites ranged from 0% to 65%. Mean naming across items was 20.1% (SD = 21.3%), which is comparable to other research (17% < M < 25%) using a similar design with famous or non-famous stimuli (Brace et al., 2000;Davies, van der Willik, & Morrison, 2000;Frowd, Hancock, & Carson, 2004). The mean was much higher for internal-features composites (M = 35.7%, ...
Recognition following long delays is superior for highly attractive and highly unattractive faces (cf. medium-attractive faces). In the current work, we investigated participants’ ability to recreate from memory faces of low, medium, and high physical attractiveness. In Experiment 1, participants constructed composites of familiar (celebrity) faces using the holistic EvoFIT system. When controlling for other variables that may influence face recognition (memorability, familiarity, likeability, and age), correct naming and ratings of likeness were superior for composites of low attractiveness targets. Experiment 2 replicated this design using the feature-based PRO-fit system, revealing superiority (by composite naming and ratings of likeness) for medium attractiveness. In Experiment 3, participants constructed composites of unfamiliar faces after a forensically relevant delay of 1 day. Using ratings of likeness as a measure of composite effectiveness, these same effects were observed for EvoFIT and PRO-fit. The work demonstrates the importance of attractiveness for method of composite face construction.
... As would be expected in general for well-known famous faces, mean rating was high for memorability (M = 3.8) and likeability (M = 3.4). Table 1 about here Individual rating scores (N = 1152) were analysed using Generalized Estimating Equations (GEE), a regression-type approach that is statistically more powerful than the popular ANOVA and provides a combined by-subjects and by-items model appropriate for the repeated observations of four sets of Likert ratings for each target photograph (Barnett, Koper, Dobson, Schmiegelow, & Manseau, 2009). The model used was full-factorial with an ordinal logistic response function and, based on homogenous correlations between elements, an Exchangeable Working Correlation Matrix. ...
Recognition following long delays is superior for highly attractive and highly
unattractive faces (cf. medium-attractive faces). In the current work, we
investigated participants’ ability to recreate from memory faces of low-, medium- and
high physical attractiveness. In Experiment 1, participants constructed composites of
familiar (celebrity) faces using the holistic EvoFIT system. When controlling for other
variables that may influence face recognition (memorability, familiarity, likeability and
age), correct naming and ratings of likeness were superior for composites of low
attractiveness targets. Experiment 2 replicated this design using the feature based
PRO-fit system, revealing superiority (by composite naming and ratings of
likeness) for medium attractiveness. In Experiment 3, participants constructed
composites of unfamiliar faces after a forensically-relevant delay of 1 day. Using
ratings of likeness as a measure of composite effectiveness, these same effects were
observed for EvoFIT and PRO-fit. The work demonstrates the importance of
attractiveness for method of composite face construction.
... Naming data were analyzed using generalized estimating equations (GEE), a multiple regression based analysis that is appropriate for correlated data (Barnett, Koper, Dobson, Schmiegelow, & Manseau, 2010;Hanley, Negassa, Edwardes, & Forrester, 2003)-to accommodate both multiple responses per participant as well as the withinsubjects factor, view at naming. The approach creates what is essentially a combined by-participants and by-items analysis of the data (for a similar analysis, see Frowd et al., 2013). ...
We investigated the impact of congruency between the witness interview and method used to construct a composite face. Experiment 1, using a typical feature-by-feature composite method, revealed that aligning cognitive processes during interview and face construction enhanced the effectiveness of composites compared with composites produced following unaligned (incongruent) procedures. Experiment 2 revealed that incorporating character judgments in the witness interview substantially enhanced identification of feature-based composites when constructing the central (internal) features first, suggesting that such judgments focus attention on this region of the face. Experiment 3 explored alignment of processes using an approach based on an evolutionary algorithm, a method requiring witnesses to create a composite by selecting from arrays based on the eye-region. A combination of character judgments, first for the whole face and then for the eye region, led to best-identified composites. Overall, results indicate that more effective composites are produced when both interview and construction procedures are aligned cognitively. Results are discussed with relevance to the theory of transfer-appropriate processing (Morris, Bransford, & Franks, 1977). (PsycINFO Database Record (c) 2019 APA, all rights reserved).
... 27 With generalised linear mixed modelling, identifying the proper covariance structure that best describes the level of correlation between data points is crucial for analysis. 28 A model of the proposed covariance structure among respondents in the OHC study is proposed in figure 2. 29 In this model, r corresponds to the correlation of individuals recruited by the same respondent and s accounts for correlation within the same cluster. This proposed covariance structure was used, but sometimes failed to converge due to sparse cells. ...
Objective
This study explores the relationship between health access barriers and diabetes in an urban First Nations population in Canada.
Design
Data from a self-identified urban First Nations population were collected using respondent-driven sampling (RDS). As no clear approach for regression modelling of RDS data is available, two logistic regression modelling approaches, including survey-based logistic and generalised linear mixed models, were used to explore the relationship between diabetes and health barriers of interest, including access to healthcare, food, housing and socioeconomic factors.
Setting
Hamilton, Ontario, Canada.
Participants
This cross-sectional study used data collected from the Our Health Counts study, in partnership with the De dwa da dehs nye>s Aboriginal Health Centre, which recruited 554 First Nations adults living in Hamilton using RDS.
Results
After adjusting for covariates, multivariable regression techniques showed a statistically significant relationship between a self-reported diagnosis of diabetes and a lack of culturally appropriate care among urban First Nations peoples (OR: 12.70, 95% CI 2.52 to 57.91). There was also a trend towards a relationship between diabetes and not having a doctor available in the area, feeling that healthcare provided was inadequate and a lack of available healthcare services in the area.
Conclusions
Urban First Nations peoples who felt the health service they received was not culturally appropriate were more likely to have diabetes, compared with those who did not feel the service they received was culturally inappropriate. Establishing more healthcare services that integrate First Nations cultures and traditions could improve access to care and the course of treatment for urban First Nations peoples living with diabetes.
... Generalized Estimating Equations (GEE) were employed, a type of regression model suitable for our dichotomous (0 and 1) responses with repeated measurements (e.g. [3]). The model was full-factorial using a binary-logistic function; a covariance structure was chosen based on equal correlations for the within-subjects factor. ...
Research has indicated that traditional methods for accessing facial memories usually yield unidentifiable images. Recent research, however, has made important improvements in this area to the witness interview, method used for constructing the face and recognition of finished composites. Here, we investigated whether three of these improvements would produce even-more recognisable images when used in conjunction with each other. The techniques are holistic in nature: they involve processes which operate on an entire face. Forty participants first inspected an unfamiliar target face. Nominally 24h later, they were interviewed using a standard type of cognitive interview (CI) to recall the appearance of the target, or an enhanced 'holistic' interview where the CI was followed by procedures for focussing on the target's character. Participants then constructed a composite using EvoFIT, a recognition-type system that requires repeatedly selecting items from face arrays, with 'breeding', to 'evolve' a composite. They either saw faces in these arrays with blurred external features, or an enhanced method where these faces were presented with masked external features. Then, further participants attempted to name the composites, first by looking at the face front-on, the normal method, and then for a second time by looking at the face side-on, which research demonstrates facilitates recognition. All techniques improved correct naming on their own, but together promoted highly-recognisable composites with mean naming at 74% correct. The implication is that these techniques, if used together by practitioners, should substantially increase the detection of suspects using this forensic method of person identification.
