1949
DOI: 10.1007/bf02289153
|View full text |Cite
|
Sign up to set email alerts
|

Variation of the standard error of measurement

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

3
35
0
3

Year Published

1950
1950
2018
2018

Publication Types

Select...
4
4
1

Relationship

0
9

Authors

Journals

citations
Cited by 70 publications
(44 citation statements)
references
References 3 publications
3
35
0
3
Order By: Relevance
“…The effect of the item characteristic curve here, considered as a transformation imposed on the frequency distribution of ability, is to squeeze together the cases that are near the middle of the ability distribution, thus producing a leptokurtic true score distribution. Empirical evidence confirming this result is found in (8,10).…”
Section: Fig 9 Derivation Of the Distribution Of True Scores For A supporting
confidence: 72%
See 1 more Smart Citation
“…The effect of the item characteristic curve here, considered as a transformation imposed on the frequency distribution of ability, is to squeeze together the cases that are near the middle of the ability distribution, thus producing a leptokurtic true score distribution. Empirical evidence confirming this result is found in (8,10).…”
Section: Fig 9 Derivation Of the Distribution Of True Scores For A supporting
confidence: 72%
“…The effect is not great for most tests in actual practice; however, for a number of years Dr. L. R Tucker has pointed out that the numerous test score distributions that he has occasion to see are usually platykurtic. Confirmatory empirical results are contained in papers by Mollenkopf (10) and Lord (8). Figure 9 may be thought of as showing the true score distribution of a test that has rather low discriminating power.…”
Section: Fig 8 Derivation Of the Distribution Of True Scores For A mentioning
confidence: 83%
“…Asumiendo que la variabilidad del error no es necesariamente constante en los diferentes niveles de puntuación de los instrumentos de medición (Fedlt y Brennan, 1989), para el BSSS8 se examinó esta variación de error a lo largo del puntaje, mediante el error estándar de medición condicional (Conditional Standard Error of Measurement, CSEM;Feldt y Brennan, 1989). Éste se calculó con el método de regresión polinómica de Mollenkopf (1949), que requiere obtener dos mitades equivalentes del instrumento. Estas mitades se formaron por el procedimiento par-impar, desde los ítems ordenados por sus respuestas promedio.…”
Section: Procedimientounclassified
“…Ciertamente, la historia de la fiabilidad puede equipararse en gran medida a la historia de la TCT. Si bien siempre han existido iniciativas para la estimación del error de medida intraindividual (Lord, 1955;Mollenkopf, 1949;Thorndike, 1951) estas no terminaron de implantarse de forma generalizada hasta la aparición de la Teoría de Respuesta al Ítem (Hambleton & Swaminathan, 1985;Lord, 1980). La TRI ofrece un complemento a la TCT que asume que los errores de medida varían en función del nivel de competencia (habilidad) estimado.…”
Section: Fiabilidad Y Validezunclassified