Robustness to non-normality of common tests for the many-sample location problem

Khan, Anisur Rahman; Rayner, G.D.

doi:10.1155/s1173912603000178

Cited by 115 publications

(63 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Within groups, the index data were not normally distributed, frequently showing skewed or multimodal distributions, due to the presence of green and dying vegetation in oiled pixels (Figure 3). Parametric tests can be robust for skewed data and moderate deviations from normality, but are affected by kurtosis [56]. The Kruskal-Wallis test [55] is insensitive to both skewness and kurtosis [56] and can be used as a measure of the difference between group medians as long as the two distributions being compared have similar variances and distributions [57].…”

Section: Comparison Methods For Cumulative Impact Of Oil and Hurricanmentioning

confidence: 99%

“…Parametric tests can be robust for skewed data and moderate deviations from normality, but are affected by kurtosis [56]. The Kruskal-Wallis test [55] is insensitive to both skewness and kurtosis [56] and can be used as a measure of the difference between group medians as long as the two distributions being compared have similar variances and distributions [57]. Figure 3 shows that this is true for our between-group comparisons.…”

Section: Comparison Methods For Cumulative Impact Of Oil and Hurricanmentioning

confidence: 99%

See 1 more Smart Citation

Marsh Loss Due to Cumulative Impacts of Hurricane Isaac and the Deepwater Horizon Oil Spill in Louisiana

et al. 2017

View full text Add to dashboard Cite

Abstract:Coastal ecosystems are greatly endangered due to anthropogenic development and climate change. Multiple disturbances may erode the ability of a system to recover from stress if there is little time between disturbance events. We evaluated the ability of the saltmarshes in Barataria Bay, Louisiana, USA, to recover from two successive disturbances, the DeepWater Horizon oil spill in 2010 and Hurricane Isaac in 2012. We measured recovery using vegetation indices and land cover change metrics. We found that after the hurricane, land loss along oiled shorelines was 17.8%, while along oil-free shorelines, it was 13.6% within the first 7 m. At a distance of 7-14 m, land loss from oiled regions was 11.6%, but only 6.3% in oil-free regions. We found no differences in vulnerability to land loss between narrow and wide shorelines; however, vegetation in narrow sites was significantly more stressed, potentially leading to future land loss. Treated oiled regions also lost more land due to the hurricane than untreated regions. These results suggest that ecosystem recovery after the two disturbances is compromised, as the observed high rates of land loss may prevent salt marsh from establishing in the same areas where it existed prior to the oil spill.

show abstract

Section: Comparison Methods For Cumulative Impact Of Oil and Hurricanmentioning

confidence: 99%

Section: Comparison Methods For Cumulative Impact Of Oil and Hurricanmentioning

confidence: 99%

Marsh Loss Due to Cumulative Impacts of Hurricane Isaac and the Deepwater Horizon Oil Spill in Louisiana

et al. 2017

View full text Add to dashboard Cite

show abstract

“…Estas dos pruebas son robustas, de forma que garantizan la certeza de sus resultados, incluso con desviaciones al supuesto de normalidad, tal como se presentó en algunas de las variables analizadas en este estudio, aunque debe tenerse precaución con las inferencias realizadas. Ambas pruebas se consideran como la mejor opción para el análisis de datos, incluso con grupos de n < 5 frente a pruebas no paramétricas (Khan & Rayner, 2003;De Winter, 2013 …”

Section: Análisis De Datosunclassified

Memoria espacial de largo plazo en humanos entrenados en un laberinto virtual

et al. 2017

View full text Add to dashboard Cite

ResumenEn este estudio se evaluó la memoria espacial a largo plazo en humanos. Para ello, se empleó un diseño cuasiexperimental en el que se entrenó a tres grupos de estudiantes de pregrado en un laberinto virtual de agua para localizar una plataforma oculta cuya ubicación era señalada por un conjunto de claves. Se realizó un pretest sin plataforma antes del entrenamiento, e inmediatamente después se condujo un postest (Grupo 0 h), así como después de un intervalo de retención de dos días (Grupo 48 h) y siete días (Grupo 168 h). En el pretest no se encontró evidencia de preferencia por alguna zona del laberinto. A lo largo de los ensayos de entrenamiento, el tiempo para encontrar la meta disminuyó sin diferencias entre grupos. Durante el postest, todos los grupos mostraron una preferencia por el cuadrante reforzado, sin embargo, el tiempo de permanencia, la distancia de nado y la precisión de la conducta de búsqueda en dicha zona fue equivalente entre el Grupo 0 h y el Grupo 48 h, aunque mayor a la mostrada por el Grupo 168 h. Estos datos indican cambios ocurridos 48 h después de la adquisición en la memoria espacial a largo plazo en humanos. Se discuten los resultados a partir de procesos generales de memoria y procesos específicos propuestos por teorías particulares de memoria espacial; y al final se abordan las implicaciones clínicas y pertinentes al campo de la psicología comparada. Palabras clave: Laberinto virtual de agua, retención, olvido espontáneo, memoria espacial. Long-term spatial memory in humans trained in a virtual maze AbstractIn this study we evaluated the long-term spatial memory in humans. A quasiexperimental design was used in which three groups of undergraduate students were trained in a virtual water maze to locate a hidden platform whose location was indicated by a set of cues. A pre-test without platform was performed prior to the training, and a post-test was conducted immediately after this (Group 0h), or after a retention interval of two (Group 48h) or seven days (Group 168h). For the pre-test, there was no evidence of preference for any area of the maze. Throughout the training trials, the time to find the goal decreased without differences between groups. During the post-test, all groups showed a preference for the reinforced quadrant, although the spent time, swimming distance, and accuracy of the search behavior in that area was equivalent between Group 0 h and Group 48 h, but higher than that shown by the Group 168 h. These data indicate changes in long-term spatial memory in humans, occurring after an interval of 48 h after its acquisition. The results are discussed on the basis of general memory processes and specific processes proposed by particular spatial memory theories. The clinical and comparative psychology implications are also addressed.

show abstract

“…Valores de η 2 < .015 reflejan un tamaño de efecto trivial, de .015 a .124 pequeño, de .125 a .254 mediano, ≥ .255 grande (Ellis, 2010). Se optó por una prueba paramétrica al ser robusta al incumplimiento de normalidad con variable numérica con distribución acampanada sin apuntamiento excesivo (Khan & Rayner, 2003).…”

Section: Análisis De Datosunclassified

Diferencias en satisfacción con la relación entre variables demográficas y religiosidad.

Rubia

2015

Salud soc.

View full text Add to dashboard Cite

RESUMENINTRODUCCIÓN: La tendencia central de la satisfacción con la relación y su asociación con variables poblacionales se suelen estimar usando muestreos no probabilísticos. OBJETIVOS: describir la distribución de la Escala de Valoración de la Relación (RAS) de Hendrick en población general de la ciudad de Monterrey, México; estudiar su asociación con variables demográficas y de religiosidad; especificar un modelo predictivo; y contrastar su invarianza entre ambos sexos. METODO: Se empleó un muestreo probabilístico, usando el método de rutas aleatorias, y se aplicó la RAS a 431 mujeres y 376 hombres. RESULTADOS: La distribución de la RAS presentó asimetría negativa. La RAS mostró diferencia entre personas casadas y en unión libre, pero no entre ambos sexos; correlacionó con el número de hijos, convicción religiosa, práctica religiosa, escolaridad, ingresos económicos y estatus laboral; y fue independiente de la edad y años de relación. Con buen ajuste e invarianza entre ambos sexos, estar casado, menor número de hijos y mayor práctica religiosa predijeron mayor satisfacción con la relación. CONCLUSION: la escala debe ser baremada por cuantiles, requiriendo baremos diferenciales por estado civil, pero no por sexo. Un modelo integrado sólo por variables demográficas y de religiosidad tiene capacidad predictiva pequeña.PALABRAS CLAVES: Estado civil, religiosidad, satisfacción con la relación, sexo.ABSTRACT INTRODUCTION: Central tendency of satisfaction with the relationship and its association with population variables tend to be measured by means of nonprobability sampling. OBJECTIVES: 1. Describe the distribution of Hendrick's RAS in the general population of Monterrey city, Mexico. 2. Study the association of such distribution across demographic variables and religiosity. 3. Specify a predicting model for satisfaction with the relationship. 4. Contrast its invariance in both sexes. METHOD: A probability sampling was employed, using the random routes method. Also, the RAS was applied to 431 women and 376 men. RESULTS: The RAS distribution resulted in negative asymmetry. This scale registered differences between married couples and non-married couples, but not between sexes. It correlated with number of children, religious orientation, religious practice, school level, economic income and occupational status, but was independent from age and years of relationship. With invariance between both sexes, being married, less number of children and more religious practice predicted higher satisfaction with the relationship. CONCLUSION: The scale should be divided by quantiles, requiring different scales for marital status but not for sex. An integrated model consisting only of demographic variables and religiosity has a small predicting capacity.

show abstract

Robustness to non-normality of common tests for the many-sample location problem

Cited by 115 publications

References 13 publications

Marsh Loss Due to Cumulative Impacts of Hurricane Isaac and the Deepwater Horizon Oil Spill in Louisiana

Marsh Loss Due to Cumulative Impacts of Hurricane Isaac and the Deepwater Horizon Oil Spill in Louisiana

Memoria espacial de largo plazo en humanos entrenados en un laberinto virtual

Diferencias en satisfacción con la relación entre variables demográficas y religiosidad.

Contact Info

Product

Resources

About