UNIDAD DE INVESTIGACIÓN:
DISEÑO Y ANÁLISIS EN PSICOLOGÍA APLICADA
(Designs in Applied Psychology)
Directora: Dolores
Frías-Navarro
Universitat de València
ÚLTIMA
ACTUALIZACIÓN 27 DE ENERO DE 2015
TAMAÑO DEL EFECTO
|
|
CONCEPTO Ø ¿Qué es un tamaño del efecto?
Robert Coe. Centre for
Evaluation & Monitoring
(CEM). Ø
Concepto. Lee A. Becker.
University of Colorado. PROGRAMAS
DE CÁLCULO Ø Descarga un programa
que ejecuta el cálculo del tamaño del efecto en Excel 5/95 y sus intervalos
de confianza. Robert Coe. Centre for Evaluation & Monitoring
(CEM). Ø Effect Size Calculators. Calcula on
line el tamaño del efecto d
de Cohen y el de correlación rYl utilizando
medias y desviaciones típicas y valores t para grupos independientes con
sus grados de libertad. Lee A.
Becker. University of Colorado. Ø
Effect Size Calculators. Calculate a standardized mean difference (d) (Cohen, Glas
and Hedges) and the strength of association (r and r2). Ø
Computation of Effect
Sizes. Comparison of groups with equal size (Cohen's d and Glass Δ), Comparison
of groups with different sample size (Cohen's d, Hedges' g) Effect
size for mean differences of groups with unequal sample size within a
pre-post design (dcorr sensu Klauer, 2001), Calculation
of d and r from the test statistics of dependent and independent t-tests, Computation
of d from the F-value of Analyses of Variance (ANOVA), Calculation of effect
sizes from ANOVAs with multiple groups, based on group means, Increase of
intervention success: The Binomial Effect Size Display (BESD) and Number
Needed to Treat (NNT), Risk Ratio, Odds Ratio and Risk Difference, Effect
size for the difference between two correlations, Computation of the pooled
standard deviation, Transformation of the effect sizes d, r, f,
Odds Ratio and η2, Computation of the effect
sizes d, r and η2 from χ2-
and z test statistics, Table of interpretation for different effect
sizes. |
|
EFFECT
SIZE CALCULATION (estimación) Whereas statistical significance tests assess the reliability of the
relationship between independent and dependent variables, effect sizes assess the strength
of the relationship. In general, you need to know the effect size you hope to
achieve to calculate statistical power. Effect size can be measured as the standardized difference between two
means,
or as the correlation between the independent variable classification and the
individual scores on the dependent variable, referred to as the effect size
correlation. Cohen's d Cohen defined d as the
difference between the means, M1-M2, divided by the
standard deviation of either group. For example, the groups in your study could
refer to the experimental and control groups. The standard deviation of
either group in your study can be used when the variances of the two groups
are homogeneous. Effect sizes are generally defined as small (d = 0.2), medium (d =
0.5), and large (d = 0.8). Several formulas could be used to calculate effect size. According to
Cohen: d = M1-M2
/ Ö
[( s1² + s 2²) / 2] d = M1-M2 /
s,
where s = Ö [å (X - M)² / N] In this case X is the raw score, M is the mean, and N is the number of
cases. F-test ANOVA Effect Size For ANOVA, the effect size index f is used. You can
compute the effect size index from the group means. Effect sizes are generally
defined as small (f = 0.1), medium (f = 0.25), and large (f = 0.4). Correlations Effect Size To evaluate the null hypothesis that a product moment correlation in
the population is zero (r = 0). The effect size symbol is r. Effect sizes are
generally defined as small (r = 0.1), medium (r = 0.3), and large (r = 0.5). Regression Effect Size Effect size for regression reflects the variance accounted for by some
source in the population relative to the residual variance proportion (η2). Effect sizes are generally defined as small (η2
= 0.02), medium (η2 = 0.15), and large (η2 =
0.35). |
|
Documentación sobre tamaño del efecto y análisis
estadístico ·
·
Online Statistics Education: An Interactive Multimedia
Course of Study TAMAÑO
DEL EFECTO: EJEMPLOS DE APLICACIONES ·
Presentación de un ejemplo: Kirsch, I., & Sapirstein,
G. (1998). Listening to Prozac but Hearing Placebo: A
Meta-Analysis of Antidepressant Medication. Prevention & Treatment, Volume
1, Article 0002a, posted June 26, 1998. |
