2004
DOI: 10.1007/bf02289858
|View full text |Cite
|
Sign up to set email alerts
|

The greatest lower bound to the reliability of a test and the hypothesis of unidimensionality

Abstract: Reliability, congeneric test, unidimensionality of a test,

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

2
214
1
19

Year Published

2005
2005
2018
2018

Publication Types

Select...
6
4

Relationship

0
10

Authors

Journals

citations
Cited by 286 publications
(236 citation statements)
references
References 32 publications
2
214
1
19
Order By: Relevance
“…This method is appropriate for ordinal data when underlying normality is not assumed. It is also well suited when the sample size is large and the number of items in a scale is small (Ten Berge and Socan, 2004) and estimates the reliability of a test divided into two parts of unequal length (Feldt and Charter, 2003). The resulting coefficients computed from age 1½ to 5 years were .…”
Section: Disruptive Behaviors In the Preschool Yearsmentioning
confidence: 99%
“…This method is appropriate for ordinal data when underlying normality is not assumed. It is also well suited when the sample size is large and the number of items in a scale is small (Ten Berge and Socan, 2004) and estimates the reliability of a test divided into two parts of unequal length (Feldt and Charter, 2003). The resulting coefficients computed from age 1½ to 5 years were .…”
Section: Disruptive Behaviors In the Preschool Yearsmentioning
confidence: 99%
“…The limitations and problems of coefficient alpha (Cronbach, 1951) as a measure of reliability estimate is well established in the literature (Sijtsma, 2009;Teo and Fan, 2013), but this index is still widely used. Therefore, the greatest lower bound (glb) to reliability (Jackson and Agunwamba, 1977;Woodhouse and Jackson, 1977) was calculated and reported as a better estimate of reliability (Ten Berge and Sočan, 2004;Sijtsma, 2009). Greatest lower bound estimates were computed by means of the FACTOR program (Lorenzo-Seva and Ferrando, 2006), which can be downloaded from http://psico.fcep.urv.es/utilitats/ factor/Download.html.…”
Section: Confirmatory Factor Analysismentioning
confidence: 99%
“…Para a retenção do número de fatores, recorrer-se-á ao método Kaiser (soma dos quadrados das cargas), com base em uma matriz de correlação policórica, haja vista a melhor adequação deste método para estimativa de variáveis latentes a partir de variáveis ordinais 31,32 . Ainda, vai-se calcular a precisão de cada fator (cada dimensão) pelos vieses do Alpha padronizado 33 , do Ômega 34 e do método GLB 35 , todos com base em matrizes policóricas, pois estas têm sido consideradas as mais adequadas para cálculos de estrutura interna com escalas ordinais de medida 36 . Conforme recomendado 30 , o modelo obtido na AFE será testado na AFC a partir dos seguintes índices de ajustes: WLSMV χ 2 , df, χ 2 /df, CFI, TLI e RMSEA.…”
Section: Procedimentos Estatísticosunclassified